1) The low and high dose both have stastically significant results
2) They will apply for the Low dose
3) The IPO numbers are much improved from the prior trial (33% vs. 21.6% for high dose and 33% vs 16.3% for low dose)
4) Most IPO controlled w/ drops and only 7.4% high dose and 3.5% low dose requring surgery. FYI, this is very low by ocular surgery standards.
5) Even patients with large loss >49 letters had very significant results.
6) Some patients exited protocl and got other treatments which caused some of the statistical issues. One might guess that those that got the sham device sought treatment, this may have depleted the comparative pool.
7) ICH Guidance E9 (FDA adotped these) would say you should use the combine data so I think there is clearly a solid rational for the larger data.
8) I am very impressed that the patients had continued improvment in BCV at 24 months over 18 months...
This is what the shorts don't want you to read....
In December 2009, we received the month 24 clinical readout for our FAME Study and have analyzed the full data set consistent with the recommendations regarding the appropriate population for primary analysis as described in the FDA-adopted International Conference on Harmonization of Technical Requirements for Registration of Pharmaceuticals for Human Use (ICH) Guidance E9, “Statistical Principles for Clinical Trials.” ICH is a joint initiative involving regulatory authorities and pharmaceutical industry representatives from Europe, Japan and the United States who discuss scientific and technical aspects of product registration.